Pose Detection of Person in a Vehicle

Abstract

A computerized method of pose detection of a person in a vehicle includes receiving, from an on-board camera of the vehicle, an image of an interior of the vehicle showing a seat of the vehicle occupied by the person. The computerized method includes obtaining at least one first characteristic of a first face bounding area and a first body bounding area associated with the occupied seat of the vehicle. The computerized method includes determining a second body bounding area and an associated second face bounding area of the person from the image. The computerized method includes determining at least one second characteristic of the second face bounding area and the second body bounding area. The computerized method includes determining a pose of the person based on the at least one second characteristic and on the at least one first characteristic.

1 . A computerized method of pose detection of a person in a vehicle, the method comprising:
 receiving, from an on-board camera of the vehicle, an image of an interior of the vehicle showing a seat of the vehicle occupied by the person;   obtaining at least one first characteristic of a first face bounding area and a first body bounding area associated with the occupied seat of the vehicle;   determining a second body bounding area and an associated second face bounding area of the person from the image;   determining at least one second characteristic of the second face bounding area and the second body bounding area; and   determining a pose of the person based on the at least one second characteristic and on the at least one first characteristic.   
     
     
         2 . The method of  claim 1  wherein obtaining the at least one first characteristic includes retrieving the first face bounding area and the first body bounding area from a first memory based on a facial identification of the person. 
     
     
         3 . The method of  claim 1  wherein:
 obtaining the at least one first characteristic includes retrieving the first face bounding area and the first body bounding area from a second memory; and 
 the first face bounding area and the first body bounding area are determined based on a mean position and/or mean area of face bounding areas and body bounding areas historically captured for persons on the seat. 
 
     
     
         4 . The method of  claim 1  wherein obtaining the at least one first characteristic includes determining the first face bounding area and the first body bounding area based on a mean position and/or mean area of face bounding areas and body bounding areas captured on a plurality of images of the interior of the vehicle for the person on the seat in a certain initialization period. 
     
     
         5 . The method of  claim 1  wherein:
 the at least one characteristic includes a position of a face bounding area and a position of a body bounding area; and 
 determining the pose of the person includes:
 calculating a left-right-leaning value of the person based on the position of the first face bounding area and the first body bounding area and the position of the second face bounding area and the second body bounding area, and 
 determining that the person is leaning right or left based on the left-right-leaning value. 
 
 
     
     
         6 . The method of  claim 1  wherein:
 the at least one characteristic includes an area covered by a face bounding area and an area covered by a body bounding area; and 
 determining the pose of the person includes:
 calculating a forward-backward-leaning value of the person based on at least two of the area of the first face bounding area, the area of the first body bounding area, the area of the second face bounding area, and area of the second body bounding area, and 
 determining that the person is leaning forward or backward based on the forward-backward-leaning value. 
 
 
     
     
         7 . The method of  claim 1  wherein the pose detection for a seat occupancy system is used to control safety functions in the vehicle. 
     
     
         8 . The method of  claim 1  wherein:
 a graphical representation on a display within the vehicle is based on an output of a seat occupancy system, and 
 the output of the seat occupancy system is based on at least one of the pose detection, the second body bounding area, and the second face bounding area. 
 
     
     
         9 . The method of  claim 1  wherein the method is performed at a periodicity and/or in response to a person-to-seat assignment change of any of the seats in the vehicle. 
     
     
         10 . The method of  claim 1  wherein determining the pose of the person is performed on a periodic basis. 
     
     
         11 . The method of  claim 1  wherein determining the pose of the person is performed in response to detection of a person-to-seat assignment change of any seats in the vehicle. 
     
     
         12 . A seat occupancy system in a vehicle, the system comprising:
 storage configured to store executable instructions; and   a set of processors configured to execute the executable instructions, wherein the executable instructions include:
 receiving, from an on-board camera of the vehicle, an image of an interior of the vehicle showing a seat of the vehicle occupied by a person; 
 obtaining at least one first characteristic of a first face bounding area and a first body bounding area associated with the occupied seat of the vehicle; 
 determining a second body bounding area and an associated second face bounding area of the person from the image; 
 determining at least one second characteristic of the second face bounding area and the second body bounding area; and 
 determining a pose of the person based on the at least one second characteristic and on the at least one first characteristic. 
   
     
     
         13 . The seat occupancy system of  claim 12  wherein the executable instructions include updating seat assignment data in response to determining the pose of the person. 
     
     
         14 . The seat occupancy system of  claim 13  wherein updating the seat assignment data includes adapting at least one of a corner position and an edge position of a seat bounding area according to at least a second body bounding box. 
     
     
         15 . The seat occupancy system of  claim 13  wherein updating the seat assignment data includes adapting parameters associated with at least one of corners and/or edges of the seat bounding area according to at least a second body bounding box. 
     
     
         16 . A vehicle comprising:
 the seat occupancy system of  claim 12 ; and   the on-board camera.   
     
     
         17 . A non-transitory computer-readable medium comprising instructions including:
 receiving, from an on-board camera of a vehicle, an image of an interior of the vehicle showing a seat of the vehicle occupied by a person;   obtaining at least one first characteristic of a first face bounding area and a first body bounding area associated with the occupied seat of the vehicle;   determining a second body bounding area and an associated second face bounding area of the person from the image;   determining at least one second characteristic of the second face bounding area and the second body bounding area; and   determining a pose of the person based on the at least one second characteristic and on the at least one first characteristic.